How Endorsement and Awareness of Gender Stereotypes Predict Girls’ Motivation in STEM

• In Event: 2-131 - Intergroup Stereotypes about STEM and Intellectual Abilities: Interdisciplinary Evidence from Singapore, USA, and Chile

Integrative Statement

Women are underrepresented in many STEM fields, especially in computer science, where only 18% of US bachelor’s degrees go to women (NSF, 2017). Stereotypes about girls’ lower STEM ability may contribute to underrepresentation. Awareness of stereotypes may contribute to stereotype threat and girls’ lower STEM performance. Yet personal endorsement of stereotypes may have a more direct link to STEM interest than awareness. We hypothesized that girls’ endorsement of computer science-gender stereotypes would show a stronger relation than awareness to their computer science interest. We also predicted this pattern would be stronger for girls in early adolescence (6th-7th grades) compared to middle childhood (3rd-5th grades), due to older girls’ greater endorsement and awareness of stereotypes.

Participants were 332 students in 3rd-7th grades from Rhode Island public schools (a unique state with a recent initiative to promote K-12 computer science) who reported their gender as female and passed an attention check. We focused on girls due to their underrepresentation in computer science; age range 8-13 years old (Mage=11.84, SD=1.46).

Participants completed an online survey, with all items on a Likert scale from 1-6. We used computer science terms familiar to participants (e.g., “coding”). For endorsement, children rated, “how good are most girls/boys at coding?” For awareness, children rated, “how good do most people think girls/boys are at coding?” Motivation was measured with three scales: belonging (3 items, α=.77; e.g., “I feel like I belong when I do coding”); self-efficacy (2 items, α=.83; e.g., “I am good at coding”); and interest (2 items, α=.81; e.g., “I like to do coding”).

We analyzed (a) ratings for each type of stereotype (endorsement or awareness) by grade levels (3rd-5th vs. 6th-7th), Figure 1; and (b) how each type correlated with motivational measures by grade levels, Table 1. We primarily used grade level instead of age to control for stereotype exposure in school, although multiple analyses will be presented. There was a significant interaction between type of stereotype and grade level, F(1,362)=11.58, p=.001. As predicted, older girls reported higher endorsement and awareness than younger girls. In terms of endorsement, younger girls expressed the counterstereotype that girls are better than boys, while older girls did not differ from neutral. In terms of awareness, younger girls did not differ from neutral, while older girls reported significant awareness of typical stereotypes. The contrast between older girls’ endorsement and awareness was particularly strong: even girls who did not endorse stereotypes were aware that most other people endorsed them. Further analyses showed that the correlations supported our hypotheses: The more that older girls endorsed typical stereotypes, the lower their motivation (belonging, self-efficacy, and interest), but their awareness was not correlated with motivation. For younger girls, endorsement correlated negatively only with self-efficacy, and awareness was positively correlated with sense of belonging.

Belief in stereotypes that boys are better at STEM may deter American girls from interest in STEM, especially as they reach middle school. Implications for targeted interventions to counteract the negative effects of stereotypes and promote girls’ STEM interest will be discussed.

Authors

• Allison Master, University of Washington
  Presenting Author

• Sapna Cheryan, University of Washington
  Non-Presenting Author

• Andrew Meltzoff, University of Washington
  Non-Presenting Author